证据冲突条件下多传感器目标识别方法比较分析  

Comparison Analysis with Method of Multi-sensor Target Identification under Evidence Conflict

作　　者：张永利 计文平 郭卫东 

机构地区：[1]中国电子科技集团公司电子科学研究院,北京100041

出　　处：《数学的实践与认识》2015年第17期158-166,共9页Mathematics in Practice and Theory

摘　　要：对于包含雷达、红外、电子支援措施和敌我识别等传感器的综合敌我识别系统,如何对异类传感器输出的不确定性信息进行有效处理是有待解决的基本难题之一.D-S证据理论是一种处理不确定性问题的有效方法.D-S证据理论对高冲突证据合成时会出现融合结果与常理相悖的情况.目前解决该问题主要使用的方法包括改变矛盾信息分配方式的组合规则法和证据信息修正法.从这两方面着手,具体分析在证据冲突情况下不同的目标识别方法,并比较其优劣,为建立综合敌我识别系统提供系统的理论支撑.For the integrated system of friend orfoe identification consisting of radar, infrared, electronic support measure （ESM） and a device of friend or foe identification, the effective processing of uncertain information from different sensors is a fundamental problem to be solved. D-S evidence theory is a useful method to deal with uncertainty problems. D-S evidence theory involves counter-intuitive behaviors when evidence highly conflicts. There are two kinds of ways to solve the question, that is, conflict evidences combination rules and evidence belief modifying method. In this paper, the two kinds of approaches axe used to solve multi-sensor target identification, and get the comparison analysis of effectiveness and performance of information fusion method in case of high conflict between evidences. Systemic theory support is provided to set up an integrated identification system of friend or foe.

关 键 词：证据理论 加权系数 证据距离 证据冲突 目标识别 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置] TP391.4[自动化与计算机技术—控制科学与工程]